MARINE ECOLOGY PROGRESS SERIES Vol. 241: 99–112, 2002 Published October 4 Mar Ecol Prog Ser Impact of introduced seastars Asterias amurensis on survivorship of juvenile commercial bivalves Fulvia tenuicostata D. Jeff Ross1, 2,*, Craig R. Johnson1, Chad L. Hewitt2 1School of Zoology, University of Tasmania, GPO Box 252-05, Sandy Bay, Tasmania 7001, Australia 2Centre for Research on Introduced Marine Pests, CSIRO Marine Research, GPO Box 1538, Hobart, Tasmania 7001, Australia ABSTRACT: The introduction and establishment of the predatory seastar Asterias amurensis in southern Australia is considered a major threat to benthic marine communities and commercial bivalves. The impact of A. amurensis on a soft sediment assemblage in SE Tasmania was quantified, with particular attention to effects of seastar predation on the survivorship of recently settled juve- niles of the commercial bivalve Fulvia tenuicostata. In a manipulative experiment, densities of F. tenuicostata juveniles were reduced by ca. 15 fold (from 580 to 35 m–2) in the presence of seastars at background densities relative to the treatment without seastars. In a feeding survey, A. amurensis exhibited preference for F. tenuicostata following settlement of the bivalve over the period from December 1997 to January 1998, which comprised 80 and 50% of the seastar’s stomach items in Feb- ruary 1998 and April 1998, respectively. However, A. amurensis preferred a variety of other prey taxa when the abundance of F. tenuicostata was low, particularly other bivalves, gastropods and the echi- noid Echinocardium cordatum. This indicates that the seastar may potentially affect the abundance of other prey taxa. The results provide further support to the hypothesis that predation by A. amuren- sis is largely responsible for the recent decline and subsequent rarity of large bivalves in its current distribution in Tasmania. The potential of seastar predation to have wider ecosystem level effects than the short-term direct effects reported here is a major concern. KEY WORDS: Seastar · Introduced species · Predation · Impacts · Bivalves · Soft sediment · Asterias amurensis Resale or republication not permitted without written consent of the publisher INTRODUCTION 1958, Lodge 1993, Lafferty & Kuris 1994) and members of Asteroidea are among the most important predators In the late 20th century, there has been a rapid accel- in marine benthic systems (Menge 1982). In Australia, eration of biological introductions due to human-aided the introduced northern Pacific seastar Asterias amu- movement of species across and between continents rensis has been highly conspicuous as a voracious gen- (Carlton & Geller 1993, Lodge 1993, Mills et al. 1993, eralist predator in soft-sediment habitats (Grannum et Vitousek 1994, Cohen & Carlton 1998, Ruiz et al. 2000). al. 1996). It was first recorded in SE Tasmania in 1986 Introduced species most likely to have a large impact (Buttermore et al. 1994, Byrne et al. 1997) where it has on native communities are those of predators (Elton become the dominant invertebrate predator in the Der- went Estuary (Grannum et al. 1996), but now also occurs at high densities in Port Phillip Bay on mainland *Present address: Department of Zoology, University of Australia. Although other invertebrate predators are Melbourne, Melbourne, Victoria 3010, Australia. present in the Derwent Estuary, such as the native *E-mail: [email protected] seastar Coscinasterias muricata, none of them attain © Inter-Research 2002 · www.int-res.com 100 Mar Ecol Prog Ser 241: 99–112, 2002 comparable densities to A. amurensis, particularly over sheltered soft sediments in areas outside the estuary such large areas and for consistently long periods. where it is part of a small commercial fishery, it is rare Densities of the seastar recorded within the estuary (up in the sediments where seastars are abundant, despite to 46 m–2; S. Ling unpubl. data) are far greater than any the presence of numerous shell remains. Adult popula- recorded during outbreaks of the species in the Ariake tions of F. tenuicostata were decimated by the seastar Sea, Japan (Nojima et al. 1986). in experiments conducted outside the seastar’s current The key role of asteroid predation as a structuring range (Ross et al. in press). The mass settlement of F. force in benthic communities largely reflects their abil- tenuicostata provided an opportunity to test the hypo- ity to exploit a wide range of food resources and habi- thesis that seastar predation on juveniles can prevent tats (Menge 1982). Asterias species, such as A. amu- the establishment and potential recovery of adult pop- rensis, are renowned for their capacity to form massive ulations in the estuary. In this paper, we undertake a aggregations and cause enormous damage to shellfish feeding survey comparing the abundance of prey taxa grounds (reviewed by Sloan 1980). In its native range, in the sediments and in seastar stomachs before, dur- A. amurensis causes considerable damage to commer- ing and after the settlement event. We also carry out a cial shellfisheries (e.g. oysters, cockles, scallops, other manipulative experiment to assess the impact of Aster- clams; Hatanaka & Kosaka 1959, Kim 1969, Nojima et ias on the survivorship of juvenile F. tenuicostata and al. 1986). Additionally, it is an opportunistic feeder on a the abundances of other species. Overall, the com- variety of epifaunal and infaunal species including bined results of this study and those from related work molluscs, ascidians, bryzoans, sponges, crustaceans, (Ross 2001) provide the first quantitative estimates of polychaetes, fish and echinoderms (Hatanaka & Kosa- the direct impact of the seastar on native assemblages ka 1959, Fukuyama & Oliver 1985, Fukuyama 1994). and wild fisheries (Ross et al. in press). On this evidence alone, it has been anticipated that the successful establishment of A. amurensis in SE Tasma- nia has the potential to profoundly affect native ben- MATERIALS AND METHODS thic marine communities and commercial species, par- ticularly bivalves (Lockhart 1995, Grannum et al. Background. The study was undertaken at Huxleys 1996). Beach (42° 58’ S, 147° 27’ E) in Ralphs Bay, a large In contrast to the extensive research on introduced shallow embayment on the eastern side of the Der- species in freshwater and terrestrial communities, the went Estuary, southeast Tasmania, Australia, at a impacts of introduced species on native communities depth of 4 to 5 m. The sediment at this site is com- in marine and estuarine habitats have been quantified posed predominantly of very fine to medium sands in only a handful of studies (e.g. Race 1982, Brenchley (Ross 2001). The habitat type at the study site is com- & Carlton 1983, Nichols et al. 1990, Lambert et al. 1992, mon in other areas of the Derwent Estuary and Posey et al. 1993, Grosholz & Ruiz 1995, Grosholz et al. adjoining bays and estuaries. The habitat range of 2000, Ross et al. in press). In Tasmania, indirect indica- both species overlaps: Asterias occurs on sheltered silt tions of impact from observations of seastar foraging to sand and on rocky reefs at 0 to 35 m depth, while behaviour, stomach contents and estimates of feeding Fulvia occurs on sheltered silt to sand at 0 to 30 m electivity suggest the potential for considerable impact depth (Edgar 1997). The average density of Asterias of Asterias amurensis on native species, particularly amurensis recorded at this site prior to this study bivalves (Morrice 1995, Grannum et al. 1996, Lockhart (1995 to 1996) was 0.53 m–2 (±0.12) (Grannum et al. & Ritz 2001). Furthermore, live bivalves >5 to 10 mm 1996, Ross 2001). In contrast, the bivalve Fulvia are rare in areas where the seastar is now abundant. tenuicostata was absent in infaunal surveys at the site The presence in the surface sediments of the shells of during the same period, despite the presence of large adults indicates that this has not always been the numerous shells in the sediment. The vast majority of case (Lockhart 1995). The predominance of juvenile F. tenuicostata present at the site during this study molluscs, and particularly bivalves, in the diet of the were newly settled juveniles of small size (1 to 13 mm seastar (Lockhart 1995, Morrice 1995) suggests that shell length; see Fig. 3a). seastar predation on juveniles may be preventing the Manipulative experiment. Experimental design: subsequent establishment and potential recovery of The effects of seastar predation on the survival of juve- adult populations in the estuary. Over the period from niles of the commercial bivalve Fulvia tenuicostata and December 1997 to January 1998, a massive settlement the abundance of other prey taxa was examined in a of the commercial bivalve Fulvia tenuicostata occurred caging experiment conducted over a 10 wk period in Ralphs Bay in the Derwent Estuary where seastars between January and April 1998. Cages consisted of a are abundant. We recorded juveniles at 530 m–2 in late rigid (1 × 1 m base × 0.7 m high) steel frame with pro- January 1998. Although F. tenuicostata is common in truding legs (0.5 m long). The cage top and sides were Ross et al.: Seastar predation on juvenile bivalves 101 covered in stiff plastic mesh (6 mm) and the cage legs the (Ofinal) unmanipulated, caged inclusion and exclu- were driven into the sediment so that 100 to 150 mm of sion plots were sampled using the suction device as the cage sides was buried to prevent passage of large described above. predators or prey by burrowing. The treatments Samples were sieved (1.0 mm mesh) prior to fixing in included: (Oinitial) open unmanipulated plot subject to 5 to 10% buffered formalin with Rose Bengal stain, background predation (sampled at 0 wk); (Ofinal) open and then rinsed in freshwater before storing in 100% unmanipulated plot subject to background predation ethanol. For core samples, all infaunal and epifaunal (sampled at 10 wk); (–S) caged exclusion, no seastars organisms (>1 mm) were sorted and identified to the present; and (+S) caged inclusion with a single seastar lowest possible taxon.